High Resolution Quantitative Risk Analysis - An Advanced Management Tool

PresentationTypical deliverables of a classical quantitative risk analysis (or QRA) effort are individual risk and societal risk data. Individual risk within the context of such a classical QRA is defined as the frequency or probability (measured over a fixed time span) of mortality of a “hypothetical and immobile” individual, who resides at a fixed location for an infinite amount of time. The determination of this individual risk number is a rather abstract exercise that yields a number without direct practical applications other than a comparison of the safety levels across different sites. Societal risk data are normally presented in the form of an F-N curve, the slope of which is a measure of the risk aversion towards accidents with multiple fatalities. Although an F-N curve is a great tool it includes concepts that are often difficult to grasp for decision-makers. For facilities where we conduct a QRA, the classical individual risk will be interpreted as a measure of the inherent process safety of those facilities. The classical individual risk and F-N curves are calculated with the aid of industry databases for equipment failures. Numerous failure scenarios are included in the effort. Failure modes are as realistic as possible; e.g., a credit will be taken for pressure venting during fired heater explosions. Unique scenarios that were found during the Process Design hazard review and during the HazOp review as well as corrective actions are also accounted for. Corrective actions are typically in the form of a redesign where we try to design a problem out (inherent safety) or control improvements (safety- instrumented systems, external risk reduction facilities, etc.). Furthermore we will compute the “true” risk to any individual working in or near the facility under review. In order to accomplish this feat we have to determine an individual’s spatial distribution in the facility and the fractional time(s) of exposure. The complexity of this endeavor requires that certain simplifications be introduced. NOVA Chemicals has worked together with TNO2 in developing the necessary software. The final result is a high resolution QRA that allows management to compare hazards to individuals on a one-to-one basis; e.g., the furnace engineer’s risk to the furnace operator’s risk. Our initial experience has been that this “humanization” of risk, making the numbers transparent and the direct risk comparison feature are of great value to management